Tech takes the Pareto principle too far
Many commenters argue that tech and startup culture lean too heavily on the “80/20 rule,” shipping MVPs or flashy demos that validate demand but never receive the polish, reliability, and completeness users ultimately expect. Others counter that Pareto thinking is valuable when resources are limited: getting 80% of the value quickly can be better than never shipping at all, especially when product‑market fit is uncertain. The exchange contrasts domains where “good enough” is acceptable (CRUD apps, many consumer tools, some games and AI assistants) with safety‑critical or long‑lived software, where cutting corners on the “last 20%” can be dangerous or deeply unsatisfying.
Meaning and Misuse of the Pareto Principle
- Several commenters argue the article treats Pareto as a sequential “first 80% of work / last 20% of work” rule, whereas it originally describes uneven distributions (e.g., 20% of causes → 80% of effects).
- Others say the analogy to features and effort is still useful, as a heuristic about diminishing returns and prioritization.
- Multiple posts stress that “80/20” is a rough power‑law intuition, not a law of nature, and warn against using it to justify social hierarchies or fatalism.
MVPs, Vertical Slices, and Product Strategy
- Strong debate over whether a game “vertical slice” is equivalent to an MVP.
- Some say a polished, limited game slice is just one kind of MVP.
- Experienced game developers counter that MVP ≈ prototype/first playable, whereas vertical slice is production‑quality and used to validate pipelines, not markets.
- In startups, MVP is framed as testing “should we build this at all?” vs. “can we build it?”, with failures like advanced hardware (AR/VR devices) cited as over‑investing pre‑validation.
Value and Cost of the “Last 20%”
- Many agree the final polish delivers emotional satisfaction, brand differentiation, and timelessness, but is expensive.
- Some see normal employment as denying developers that completion satisfaction, reserving it for hobbies (e.g., woodworking).
- Others emphasize opportunity cost: the time to perfect one feature could ship many “good enough” features customers actually value more.
User Expectations and “Good Enough”
- Multiple comments: most users accept “passable” quality in housing, food, apps, etc.; perfection is overkill.
- Counterpoint: in crowded markets, the extra 80% of refinement on core 20% of features is exactly what creates competitive advantage.
Domains Where Pareto Fails
- Safety‑critical systems (medical devices, power plants, serious drones, flight control, some robotics) are cited as examples where you must aim for near‑perfection.
- Concern that fast‑and‑loose web/SaaS cultures are bleeding into domains like self‑driving cars and AI.
AI and 80% Reliability
- Some see current “AI” as an archetypal 80% solution: impressive demos, but 20% error rates make it hard to rely on.
- Others note that for many less‑skilled users, that 80% already exceeds their own baseline and delivers real value (e.g., writing, explanations).